cub200example

Gyanachand1 · Oct 31, 2020 · d2bec23 · d2bec23
1 parent baef08a
commit d2bec23
Showing 1 changed file with 72 additions and 0 deletions.
diff --git a/examples/cub200/upload_cub2010.py b/examples/cub200/upload_cub2010.py
@@ -0,0 +1,72 @@
+import os
+from PIL import Image
+from hub import Transform, dataset
+import numpy as np
+import scipy.io
+
+class CubGenerator(Transform):
+
+	def meta(self):
+		return {
+			"labels": {"shape": (1,), "dtype": "str", "dtag": "text"},
+			"image": {"shape": (1,), "dtype": "object", "dtag": "image"},
+			"__header__": {"shape": (1,), "dtype": "str", "dtag": "text"},
+			"__version__": {"shape": (1,), "dtype": "str", "dtag": "text"},
+			"wikipedia_url": {"shape": (1,), "dtype": "str", "dtag": "text"},
+			"seg": {"shape": (1,), "dtype": "int", "dtag": "segmentation"},
+			"bbox": {"shape": (1,), "dtype": "int32", "dtag": "box"},
+			"flickr_url": {"shape": (1,), "dtype": "str", "dtag": "text"},
+		}
+
+	def forward(self, image_info):
+		ds = {}
+		ds["labels"] = np.empty(1, dtype="str")
+		ds["labels"][0] = image_info["label"]
+
+		ds["image"] = np.empty(1, object)
+		ds["image"][0] = np.array(Image.open(image_info["image_path"]).convert('RGB'))
+
+		ds["__header__"] = np.empty(1, object)
+		ds["__header__"][0] = image_info["anno"]["__header__"]
+
+		ds["__version__"] = np.empty(1, object)
+		ds["__version__"][0] = image_info["anno"]["__version__"]
+
+		ds["wikipedia_url"] = np.empty(1, object)
+		ds["wikipedia_url"][0] = image_info["anno"]["wikipedia_url"]
+
+		ds["seg"] = np.empty(1, object)
+		ds["seg"][0] = image_info["anno"]["seg"]
+
+		ds["bbox"] = np.empty(1, object)
+		ds["bbox"][0] = image_info["anno"]["bbox"]
+
+		ds["flickr_url"] = np.empty(1, object)
+		ds["flickr_url"][0] = image_info["anno"]["flickr_url"]
+		return ds
+
+
+def load_dataset(path):
+	image_info_list = []
+	with open(os.path.join(path, "attributes", "images-dirs.txt"), "r") as fr:
+		images = [filename.split()[1] for filename in fr.readlines()]
+
+	for image in images:
+		image_info = {}
+		anno = scipy.io.loadmat(os.path.join(path, "annotations-mat", image[:-4]))
+
+		image_info["anno"] = anno
+		image_info["image_path"] = os.path.join(path, "images", image)
+		image_info["label"] = image.split('/')[0]
+		if os.path.exists(os.path.join(path, "images", image )):
+			image_info_list.append(image_info)
+
+	ds = dataset.generate(CubGenerator(), image_info_list)
+	return ds
+
+
+if __name__ == "__main__":
+	dataPath = input("Enter path to dataset : ")
+	ds = load_dataset(dataPath)
+	dataStore = input("Enter username/dataset")
+	ds.store(dataStore)